Accuracy and epistemic deference

马上就要抛硬币了。你知道这是一枚均匀的硬币。也就是说，你知道它有50%的几率正面着地。但这就是你所知道的。你对它会正面落地的信心有多大？50%,肯定。也就是说，在形成自己的观点时，你应该把它命中人头的机会视为一种专家的意见。这就是所谓的基本原则。在形成你现在的观点时，你似乎应该听从另一种专家的意见，那就是你未来的自己。如果你知道未来的自己对某一结果有70%的信心，那么你自己也应该有70%的信心，至少在正常情况下是这样。毕竟，未来的自己通常比现在更了解情况，而且需要更长的时间来权衡证据。这被称为反射原则。本研究项目的中心问题是：我们为什么要遵循本原原则和反思原则？如果我们违反了其中的任何一条，我们在哪些方面犯了错？在上面，我通过吸引我们对如何形成我们的观点的某些直觉来激发这些原则。但直觉是出了名的不可靠，而且往往因其产生方式而异。所以理想情况下，我们希望这些规范有一个更可靠、更严格的理由。这样的论证在哲学文献中相当缺乏。在这个项目中，我将研究一种可能性，通过一种新的策略来给出一个更严格的理由，这种策略在回答相关问题时已经起到了很大的作用。这个想法是这样的：考虑到世界可能是这样的，某些程度的自信比其他程度的自信更好。例如，一个对所有真理都有高度自信而对所有谎言都有非常低自信的人比对谎言有高度自信而对真理有很低自信的人有更好的观点。受此启发，一些哲学家试图定义一种衡量标准，以衡量给定的观点对世界的特定方式有多好。有了这样的尺度，我们就可以用下面的方法对“主要原则”和“反思原则”作出更严格的证明。假设这样的情况，无论何时你有一些观点违反了基本原则，那么就会有一些观点满足它你可能会有一些观点比你实际的观点更好（根据观点善良度的衡量）在任何可能的情况下。那么，这似乎就证明了“主要原则”的合理性。毕竟，不管世界是什么样子，最好还是按照这个原则说的去做。所以这种论证主要原则的方式保证了一个严格的证明，如果它有效的话。我们可能希望对反射原理也有类似的论证。它有用吗？我最初的研究表明，它确实提供了一些关于客观机会和未来自我的详细哲学主张。本研究项目旨在提供使这一论点成立所需的哲学基础。

A coin is about to be tossed. You know that it is a fair coin. That is, you know that there is a 50% chance that it will land heads. But this is all you know. How confident should you be that it will land heads? 50%, surely. That is, you should treat the chance of it landing heads as a sort of expert to whose opinion you defer when forming your own. This is known as the Principal Principle. Another sort of expert to whose opinion it seems you should defer when forming your current opinion is your future self. If you know that your future self will be 70% confident in a particular outcome, you should have 70% confidence yourself, at least in normal circumstances. After all, your future self is typically better informed than you are, and has had longer to weigh the evidence. This is known as the Reflection Principle. The central question of this research project is this: Why should we obey the Principal Principle and the Reflection Principle? In what way do we err if we violate either of them?Above, I motivated these principles by appealing to certain intuitions we have about how we should form our opinion. But intuitions are notoriously unreliable and often vary depending on how they are elicited. So ideally we would like a more reliable, rigorous justification for these norms. Such justifications have been rather lacking in the philosophical literature. In this project, I will investigate the possibility of giving a more rigorous justification by appealing to a new strategy that has been used to great effect in answering related questions. The idea is this: Given a way that the world might be, certain levels of confidence are better than others. For instance, someone who has a very high degree of confidence in all truths and a very low degree of confidence in all falsehoods has better opinions than someone who has high confidence in falsehoods and low in truths. Inspired by this, some philosophers have tried to define a measure of how good a given opinion is for a particular way the world might be. With this sort of measure in hand, we might give a more rigorous justification of the Principal Principle and the Reflection Principle in the following way. Suppose it were the case that, whenever you had some opinions that violate the Principal Principle, then there were some opinions that satisfy it that you might have instead and that are better than your actual opinions (according to the measure of goodness of opinions) given any possible way the world could be. Then, it seems, this would justify the Principal Principle. After all, it would be better to do what that principle says you should do however the world turns out to be. So this way of arguing for the Principal Principle promises a rigorous justification if it works. And we might hope for a similar argument for the Reflection Principle.Does it work? My initial research shows that it does providing certain detailed philosophical claims about objective chances and future selves can be established. This research project seeks to provide the philosophical underpinnings required to make this argument work.